It has long been recognized that cellular interactions are important modulators of genetic programs controlling differentiation. The goal of the proposed research is to identify and examine genes that mediate communication between cells during mammalian cerebellar development. This may involve direct cell-cell interaction or indirect interaction via the extracellular matrix or paracrine signals. To accomplish this, the project will study the family of CNS proteins possessing elements that resemble epidermal growth factor (EGF). The EGF motif was chosen because a number of Drosophila genes containing sequences with strong homology to EGF play a central role in the development of the fly nervous system. It is through their EGF-like regions that these genes are thought to act. In order to identify mammalian representatives of this family, a mouse cerebellar granule cell cDNA library was constructed and screened using a region of the Drosophila Notch gene encompassing its 39, non-identical EGF-like repeats. Thus far 4 cDNA clones have been obtained: NM 11, 20, 22, and 51, whose partial DNA sequences align with appropriate regions of the Notch gene and, thus, may prove to be related to Notch. Northern blot analysis indicates that 3 of these MN genes are developmentally regulated in cerebellum. These NM clones will be studied using a variety of molecular and biological approaches: 1) the structure of MN cDNAs will be characterized first by partial DNA sequencing, to establish that clones are unique. Full length cDNAs will be obtained from whose sequence structural and functional details of their encoded proteins will be inferred. 2) the temporal and spatial distribution of MN gene expression in the mouse will be determined initially by northern blot analysis and then by in situ hybridization using a variety of ages and tissues. In addition, the cellular and subcellular distribution of MN gene products will be examined by immunohistochemistry using antibodies raised against MN fusion proteins. 3) The action(s) of MN genes in cerebellar neurodevelopment will be investigated using three in vitro systems as model: a) embryonic and early postnatal cerebellar cells in dissociated culture that include mitotic, migrating and resting neurons b) embryonal carcinoma (EC) cell lines induced to differentiate into neuronal and glial cell types. c) molecular interactions of MN gene products introduced into continuous cells that do not normally synthesize MN proteins. 4) Based on data obtained in 1, 2, and 3, MN cDNAs will be selected for cloning of genomic DNA. Genomic clones will be restriction mapped in preparation for later functional in vivo experiments and in vitro transcriptional studies. Experiments pursued in this grant proposal will lead to examination of the function of selected MN genes in vivo by knocking out expression of targeted MN genes using homologous recombination methodology.